Audio Sync Problem using the big 3

[newguy007]

Hello, I tried to backup my Once Upon a Time in America, 2 disc special edition (NTSC REGION 1) with the Big 3 & Scenarist method http://www.doom9.org/mpg/ra-guide.htm the other day but I had some problems with the backup of disc 2. The audio and video are considerably out of sync for the main movie. The video plays, then after some time (don't know the exact delay; is there a way to find out?) the audio starts. The menus and everything else are fine. I've never encountered this before using the big 3 method and couldn't find any related threads. Since everything is essentially automated, I really can't tell when the problem occurs during the process. Disc 1 came out beautifully, but I don't know what's wrong with disc 2. Can I fix this with a simple edit of some files instead of starting all over, doing the entire proces again? What I really need is to find the exact time of delay, then maybe editing the ac3 file might fix it. But I hope there's at least way to fix this problem without doing another encode, so I won't have to spend an extra 12 hours.

I'm using the latest versions of the software mentioned in the guide and Scenarist v2.7.

Thank you

[quantum]

Interesting coincidence since I noticed the same problem with Once Upon a Time in America disk 2. I thought I must have done something wrong with removing a pgc in ReAuthorist. I never found an answer as I used IC7 to backup the second time.

[Eyes`Only]

Hmm did you have 'autocorrect ac3 delays' checked in DIF4U?

If so, what was the delay shown in the ac3 filename? Was it within the threshold setting you selected?

Issues such as yours cannot be diagnosed without detailed information, if at all.

[newguy007]

Yes, I followed the guide religiously.

VTS__02_P01-80-448K-[-18ms]-ch6English.AC3

It seems like Doitfast4u didn't correct the delay since 18 ms is not within range of the threshold, which was set to the default 40 ms. Do you think that is what caused the problem? The guide says "Most people begin to hear audio displacement starting at 50-60 ms." Of course if the audio delay for the authored movie was 18 ms everything would be fine and dandy. But the displacement in the authored movie was nowhere near 18 ms -- more like 5 seconds at least.

Is there a way I can find out the exact delay for the authored movie? I could do the trial-and-error way by cropping the ac3 file until I get it just right, but that would take a considerable amount of time.

If there's any other information you'd like, please let me know.

[Eyes`Only]

The easiest way to find the delay is to just play the muxed vob and watch the time on your software dvd player, using that will at least get you extremely close, and after you correct it, possibly to the point where you wouldn't notice the delay anymore.

[newguy007]

Thanks for the tip. Do you know of a player that will display the duration down to the millisecond? So far I've tried using powerdvd and ifoedit but they only display seconds as the smallest unit.

edit:

I've searched for all the players and none of them display the ms. Guess I'll have to use a stopwatch.

[ShaneZ]

Originally posted by newguy007
Thanks for the tip. Do you know of a player that will display the duration down to the millisecond? So far I've tried using powerdvd and ifoedit but they only display seconds as the smallest unit.

edit:

I've searched for all the players and none of them display the ms. Guess I'll have to use a stopwatch.

I have run into this problem on one occasion. Eyes`Only is correct as usual. The only way to do it really is play around with AC3Delay Corrector, remux the project and check to see if you got the delay set correctly. I had to do it three or four times before I got the delay correct, but the end result was right on the money. :D

[quantum]

Of far more interest to me is why this particular title has the audio sync problem. It was verified by two independent sources. We can work around it, but will it happen again? On what disks? Can the problem be fixed?

[newguy007]

Thanks for letting me know about that AC3 Delay Corrector program, shanez. That really helped. The good news is I finally made a working backup with correct sync and everything! The magic number is -16100 for the start delay, for those of you who are having similar problems.

edit:

Well... almost working. The subs are still being displayed at the time of the previous delay. Is there any way I can fix the delay on the subs too so it will be in sync with the audio and video? So essentially somewhere during the whole process, not only the audio was out of place, but the subs were too. I appreciate all your help.

[Xitrum]

I had the same problem with the PAL version of "Once Upon the time in America" and spent a lot of time trying to figured it out how to fix it. I eventually got the 2nd disk backed up similar to the original disk.
The problem with the subtitles are due to, I think, a bug in vobsub. The reason why I said that is because, the subs produced by vobsub for this movie are all over the places, in my case, the subs for the first few chapters of the second disk was missing. Initially, I thought there was something I did, but after using DIF4U to re-do the process again, I had the same results.
I then turned to subrip to do the ripping for the subtitles, at the first try of subrip, I used the .idx file produced by DIF4U. This time I got all the subs ripped (i.e. no missing subs), however, when I looked at the time info file (I think the extention is .sst) the timing was all over the place, by that I mean, the sub for chapter 3 was displayed first, and then it went back to display the sub for chapter 1, then chapter 4 and then chapter 2 etc...
Finally, I went back to the Original disc 2, DVDDecrypted out, in file mode, to the hard drive, then used suprip to rip the subtitles.
This time Subrip gave me a perfect rip for the subtitles (ie, no missing subs or out-of-order subs), however there are still some delays in the subs, so I used the utilities within subrip to "minus" 15secs from the .sst file and everything was perfect.
With the audio, there was also a delay as well, so what I did was, in Scenarist (I did not used the AC3 Delay corrector), I zoomed in on the time scale of the movie track, and highlighted the audio track and in the properties windows, under the "start time to video" which originally showed 00:00:00;00, I changed to 00:00:01;11.
After re-authored, the movie and the subs are all in sync.

The question I wanted to ask the expert is:
I did backed up a few movies that spanned over 2 disc in the past, and most of them had the similar structure in the second disk, in that, if you look at the structure in IFOEdit, at the start of the movie PGC, there are at least 20 short cells with about 25 frames in each cell, and each of there cells have the layer break next to them. Could any experts explain to me why the movie was done this way, and how to go about recreate the structure so the movie works the way it should. (without the manual workarounds) I used re-authorist to created the structure for me and the structure looked very similar to the original

[JFerguson]

An old thread, but a problem that beleaguers us still. I got the same result doing disc 2 of this title.

Here's the IFO for the movie:

VTS overview:

Menu attributes:
Video: MPEG-2 720x480 (NTSC) (NTSC 525/60) (16:9) (pan-scan)
Audio 1: Not Specified (Dolby AC-3) 2ch 48Kbps DRC
SubPicture 1: Not Specified (2-bit rle )

Title Set (Movie) attributes:
Video: MPEG-2 720x480 (NTSC) (NTSC 525/60) (16:9) (letterboxed)
Audio 1: English (Dolby AC-3) 6ch 48Kbps DRC (ID: 0x80)
Audio 2: Français (Dolby AC-3) 6ch 48Kbps DRC (ID: 0x81)
Audio 3: English (Dolby AC-3) 2ch 48Kbps DRC (ID: 0x82)
SubPicture 1: English (2-bit rle ) (ID: 0x20,0x21)
SubPicture 2: Français (2-bit rle ) (ID: 0x22,0x23,0x20)
SubPicture 3: Español (2-bit rle ) (ID: 0x24,0x25,0x20)

PGC_1 (program chain): [Title(TTN): 1] [01:49:15.25 / 30 fps] (Programs: 59) (Cells: 59) (uses VOB-IDs: 2,3,4,5)
[Ch 01] [Pg 01] [Cell 01] [V/C Id: 2/ 1] : time: 00:00:00.15 / 30 fps [Pos: 00:00:00.15]
[Ch 02] [Pg 02] [Cell 02] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:01.00]
[Ch 03] [Pg 03] [Cell 03] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:01.15]
[Ch 04] [Pg 04] [Cell 04] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:02.00]
[Ch 05] [Pg 05] [Cell 05] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:02.15]
[Ch 06] [Pg 06] [Cell 06] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:03.00]
[Ch 07] [Pg 07] [Cell 07] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:03.15]
[Ch 08] [Pg 08] [Cell 08] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:04.00]
[Ch 09] [Pg 09] [Cell 09] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:04.15]
[Ch 10] [Pg 10] [Cell 10] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:05.00]
[Ch 11] [Pg 11] [Cell 11] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:05.15]
[Ch 12] [Pg 12] [Cell 12] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:06.00]
[Ch 13] [Pg 13] [Cell 13] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:06.15]
[Ch 14] [Pg 14] [Cell 14] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:07.00]
[Ch 15] [Pg 15] [Cell 15] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:07.15]
[Ch 16] [Pg 16] [Cell 16] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:08.00]
[Ch 17] [Pg 17] [Cell 17] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:08.15]
[Ch 18] [Pg 18] [Cell 18] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:09.00]
[Ch 19] [Pg 19] [Cell 19] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:09.15]
[Ch 20] [Pg 20] [Cell 20] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:10.00]
[Ch 21] [Pg 21] [Cell 21] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:10.15]
[Ch 22] [Pg 22] [Cell 22] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:11.00]
[Ch 23] [Pg 23] [Cell 23] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:11.15]
[Ch 24] [Pg 24] [Cell 24] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:12.00]
[Ch 25] [Pg 25] [Cell 25] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:12.15]
[Ch 26] [Pg 26] [Cell 26] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:13.00]
[Ch 27] [Pg 27] [Cell 27] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:13.15]
[Ch 28] [Pg 28] [Cell 28] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:14.00]
[Ch 29] [Pg 29] [Cell 29] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:14.15]
[Ch 30] [Pg 30] [Cell 30] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:15.00]
[Ch 31] [Pg 31] [Cell 31] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:15.15]
[Ch 32] [Pg 32] [Cell 32] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:16.00]
[Ch 33] [Pg 33] [Cell 33] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:16.15]
[Ch 34] [Pg 34] [Cell 34] [V/C Id: 2/ 1] Layer Br.: time: 00:00:00.15 / 30 fps [Pos: 00:00:17.00]
[Ch 35] [Pg 35] [Cell 35] [V/C Id: 3/ 1] Layer Br.: time: 00:02:41.19 / 30 fps [Pos: 00:02:58.19]
[Ch 36] [Pg 36] [Cell 36] [V/C Id: 3/ 2] : time: 00:02:38.20 / 30 fps [Pos: 00:05:37.09]
[Ch 37] [Pg 37] [Cell 37] [V/C Id: 3/ 3] : time: 00:05:26.15 / 30 fps [Pos: 00:11:03.24]
[Ch 38] [Pg 38] [Cell 38] [V/C Id: 3/ 4] : time: 00:07:37.15 / 30 fps [Pos: 00:18:41.09]
[Ch 39] [Pg 39] [Cell 39] [V/C Id: 3/ 5] : time: 00:07:42.05 / 30 fps [Pos: 00:26:23.14]
[Ch 40] [Pg 40] [Cell 40] [V/C Id: 3/ 6] : time: 00:05:10.25 / 30 fps [Pos: 00:31:34.09]
[Ch 41] [Pg 41] [Cell 41] [V/C Id: 3/ 7] : time: 00:02:08.16 / 30 fps [Pos: 00:33:42.25]
[Ch 42] [Pg 42] [Cell 42] [V/C Id: 3/ 8] : time: 00:05:07.20 / 30 fps [Pos: 00:38:50.15]
[Ch 43] [Pg 43] [Cell 43] [V/C Id: 3/ 9] : time: 00:01:38.25 / 30 fps [Pos: 00:40:29.10]
[Ch 44] [Pg 44] [Cell 44] [V/C Id: 3/10] : time: 00:01:02.06 / 30 fps [Pos: 00:41:31.16]
[Ch 45] [Pg 45] [Cell 45] [V/C Id: 4/ 1] Layer Br.: time: 00:06:06.18 / 30 fps [Pos: 00:47:38.04]
[Ch 46] [Pg 46] [Cell 46] [V/C Id: 4/ 2] : time: 00:02:02.11 / 30 fps [Pos: 00:49:40.15]
[Ch 47] [Pg 47] [Cell 47] [V/C Id: 4/ 3] : time: 00:04:43.10 / 30 fps [Pos: 00:54:23.25]
[Ch 48] [Pg 48] [Cell 48] [V/C Id: 4/ 4] : time: 00:04:40.05 / 30 fps [Pos: 00:59:04.00]
[Ch 49] [Pg 49] [Cell 49] [V/C Id: 4/ 5] : time: 00:04:20.20 / 30 fps [Pos: 01:03:24.20]
[Ch 50] [Pg 50] [Cell 50] [V/C Id: 4/ 6] : time: 00:04:51.19 / 30 fps [Pos: 01:08:16.09]
[Ch 51] [Pg 51] [Cell 51] [V/C Id: 4/ 7] : time: 00:03:44.25 / 30 fps [Pos: 01:12:01.04]
[Ch 52] [Pg 52] [Cell 52] [V/C Id: 4/ 8] : time: 00:05:09.25 / 30 fps [Pos: 01:17:10.29]
[Ch 53] [Pg 53] [Cell 53] [V/C Id: 4/ 9] : time: 00:06:36.25 / 30 fps [Pos: 01:23:47.24]
[Ch 54] [Pg 54] [Cell 54] [V/C Id: 4/10] : time: 00:02:00.00 / 30 fps [Pos: 01:25:47.24]
[Ch 55] [Pg 55] [Cell 55] [V/C Id: 4/11] : time: 00:05:57.20 / 30 fps [Pos: 01:31:45.14]
[Ch 56] [Pg 56] [Cell 56] [V/C Id: 4/12] : time: 00:05:56.26 / 30 fps [Pos: 01:37:42.10]
[Ch 57] [Pg 57] [Cell 57] [V/C Id: 4/13] : time: 00:03:44.29 / 30 fps [Pos: 01:41:27.09]
[Ch 58] [Pg 58] [Cell 58] [V/C Id: 4/14] : time: 00:03:02.00 / 30 fps [Pos: 01:44:29.09]
[Ch 59] [Pg 59] [Cell 59] [V/C Id: 5/ 1] : time: 00:04:46.16 / 30 fps [Pos: 01:49:15.25]


Like others, I don't really like messing with delays and whatnot. The purpose of all those dummy programs is simply to continue the chapter numbering structure from the first disc (Chapters 1-34).

What throws the process off, I think, is that there are a number of jumps built into the navigation that skip over all of those dummy programs and upon playing, start right from Ch 35.

The audio (and subtitle) delay is roughly 17000 ms, which is the playing time of Ch 1-34.

I guess there's nothing there that screams "Demux by Vob ID", but I gave it a try on Ch 35. It looked good, so I did the whole disc and it appears to work.

So, I guess that's the trick?!! :)

[Eyes`Only]

Yep. 1 thru 34 are the same VobID, so it makes sense.

[JFerguson]

Is that also true, then, when an 00:00:00:15 Vob ID is used at the end of several PGCs?

[Trahald]

yeah.. i had to do that (VOBID) with The Stand. disks with the main movie having stills at the beginning of the pgcs are rare however. if you discover it too late, manually tweaking (you have to add delay since the stills have no sound) the audio delay will save you a reencode of the video (34x.15 = 17 secs or +17000 ms)

[Eyes`Only]

Originally posted by JFerguson
Is that also true, then, when an 00:00:00:15 Vob ID is used at the end of several PGCs?

Technically, yes. But in that instance, it's usually a blank still image, and a PGC demux will work just fine. You'll lose a couple of bytes since you'll be duplicating data, but it's negligible.